Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Recent Advances in Open Set Recognition: A Survey
Chuanxing Geng, Sheng-Jun Huang, Songcan Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use CD-OSR (HDP-based modeling of unknowns) instead of threshold-based W-SVM at highest openness | non-depth OSR methods · PENDIGITS (O*=18.35%) | micro-F-measure (%): 88.39 -> 98.43 | Improved |
| Threshold-based methods as openness increases (W-SVM on LETTER vs PENDIGITS) | W-SVM · LETTER and PENDIGITS | micro-F-measure (%) | Mixed |
| Increase openness from O*=0% to 25.46% for 1-vs-Set | 1-vs-Set · LETTER | micro-F-measure (%): 81.51 -> 42.08 | Hurt |
| Use C2AE (split closed-set classification and open-set identification) instead of SoftMax | deep OSR · CIFAR10 (O*=13.40%) | AUROC (%): 67.7 -> 89.5 | Improved |
| Joint classification and reconstruction (CROSR) instead of SoftMax | deep OSR · MNIST | AUROC (%): 97.8 -> 99.8 | Improved |
| Add instance generation of unknowns (OSRCI) vs C2AE | OSRCI · CIFAR+50 (O*=61.51%) | AUROC (%): 93.7 -> 82.7 | Hurt |
| OpenMax instead of SoftMax | SoftMax · CIFAR+50 (O*=61.51%) | AUROC (%): 80.5 -> 79.6 | No clear effect |
Closest papers by meaning
- Open-Set Recognition: A Good Closed-Set Classifier is All You Need (ICLR 2022 · similarity 0.92)
- Learning Network Architecture for Open-Set Recognition (AAAI 2022 · similarity 0.90)
- Sparse Representation-Based Open Set Recognition (TPAMI 2016 · similarity 0.90)
- All Beings Are Equal in Open Set Recognition (AAAI 2024 · similarity 0.89)
- Frequency Shuffling and Enhancement for Open Set Recognition (AAAI 2024 · similarity 0.88)
- From Coarse to Fine-Grained Open-Set Recognition (CVPR 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
